Localized TEC enhancements in the Southern Hemisphere

The paper is dedicated to investigation of localized TEC (total electron content) enhancements (LTEs), particularly of LTE series, detected in the Southern Hemisphere using global ionospheric maps for different solar activity years (2014, 2015, 2018). It is shown that LTE intensity varies in dependence on solar flux and does not directly depend on interplanetary magnetic field orientation. The enhancements occur in a subsolar region and could be observed during a continuous series of days. The highest LTE occurrence rate is observed during period of local winter (April-September) for all analyzed years. The longest observed LTE series was detected during 2014 and lasted 80 days or 120 days if we exclude 2 daily gaps.


Introduction
The Southern Hemisphere (SH) ionosphere has not been investigated so broadly as one of the Northern Hemisphere (NH).
Historically, most of the geophysical observations and measurements have been made to the north of the equator. Even now, having lots of observatories all around the globe, we have a lack of ground-based observations for a larger part of the Southern Hemisphere since it is mostly occupied by ocean. Satellite measurements allow us to investigate ionosphere over oceans but due to its high variability and the movement of satellites it is very difficult to observe the same region in the same conditions.
It is known that Southern Hemisphere contains some anomalous regions. South Atlantic Magnetic Anomaly (SAMA) is formed by a configuration of geomagnetic field which has a global minimum of intensity over South Atlantic and South America and makes it easier for energetic particles of inner radiation belt to precipitate, thus increasing ionospheric conductivity over the region (Abdu et al., 2005). South of the SAMA, in the south-eastern Pacific and South Atlantic Antarctic regions, combination of the geomagnetic field features and thermospheric winds produces an inverted diurnal plasma density pattern at equinoxes and in SH summer (October-March): the nighttime maximum is larger than the daytime minimum, and the phenomenon is known as the Weddell Sea Anomaly (WSA) (Horvath, 2006). Jakowski et al. (2015) showed that during periods of low solar activity in Asian longitudinal sector of SH it is possible to observe so called nighttime winter anomaly (NWA), when values of electron concentration are higher in winter than in summer. At the same time Yasyukevich et al. (2018) showed that winter anomaly manifests itself much less intensively in SH that in NH. It is 1 5 10 15 20 quite clear that due to these anomalies the structure and dynamics of ionosphere in both hemispheres should be different and should be investigated separately.
The most widely used and generally accepted the International Reference Ionosphere (IRI) empirical model (e.g., Bilitza, 2018) does not predict some features of the SH ionosphere sufficiently. Karia et al. (2019) analyzing predictions of IRI-2016 showed that the model does reproduce the observed NWA effect, though at a different longitude and could be improved for better predictions. Comparing TEC measurements and results of IRI-PLAS, Alcay and Oztan (2019) found that in SH the model generally overestimates the GPS-TEC measured at stand-alone stations with the maximal difference about 15 TECU. Karpachev and Klimenko (2018) proposed a new model reproducing the structure of the high-latitude ionosphere more accurately than IRI-2016 and noted that inaccuracies of IRI in that region are connected with inaccuracy of ground-based sounding data, which varies during a day. However, none of these models predict the occurrence of localized enhancements of electron concentration especially in Southern Hemisphere.
The most typical irregularities in distribution of electron concentration are produced during geomagnetic storms. Foster and Coster (2007) investigating storm enhanced densities (SED). They showed that during severe and extreme storms it is possible to detect SEDs which in maps of total electron content (TEC) could be observed as localized TEC enhancements (LTE). The authors showed that during a storm recovery phase LTEs can be detected in the night side ionosphere at the middle latitudes of both hemispheres, in magneto-conjugated regions. The authors note that the observed enhancements are approximately corotating in place over the positions in which they were formed earlier in the event. However, the LTE phenomenon studied by Foster and Coster (2007) is different from the LTE phenomenon studied by us. During analysis of ionospheric response to geomagnetic storm of 15 August 2015 Edemskiy et al. (2018) detected a curious LTE in global ionospheric maps (GIMs). Unlike the LTEs observed by Foster and Coster (2007) this enhancement was observed in sunlit (near-noon) area of the Southern Hemisphere and lasted for several hours. It was not corotating but changing position following the Sun and propagating along the geomagnetic parallels. Using quite a simple detection algorithm Edemskiy et al. (2018) found about 30 similar events in the Southern Hemisphere during 2010-2016 and the most of the detected LTEs were observed during relatively disturbed periods. The authors showed direct dependence of number of the detected LTEs on solar activity level and suggested that the generation of the enhancements is connected with the orientation of interplanetary magnetic field (IMF), namely with Bz.
The present article is an attempt to detect more LTEs developing in Southern Hemisphere during different solar activity periods and to investigate them more carefully trying to understand mechanisms of their generation. Section 2 describes data and methods, section 3 presents results, section 4 deals with discussion and possible mechanism, and section 5 summarizes main results.

Data and methods
The algorithm used by Edemskiy et al. (2018) had some disadvantages. The used fixed detection threshold did not allowed them to detect relatively weak LTEs. The applied comparison with a weekly TEC median excluded from the consideration 2 30 35 40 45 50 55 possible series of such formations. Trying to improve the effectiveness of LTE detection we introduced following criteria for the TEC formation. In this paper a TEC enhancement is considered as LTE if it is: -located in middle latitudes of sunlit region. Mainly we investigate LTEs, which are clearly observed in Indian and Southern part of Atlantic Oceans and do not take into account enhancements in Northern Hemisphere. At the same time, LTEs in SH are not accompanied by any LTE in NH and such a focusing on SH LTEs is quite reasonable.
-spatially limited by relatively lower TEC values. Normalized difference between squared maximal value in LTE and minimal one at its border (Δ=1 -(Iedge/Imax)^2) should be no less than 20%. Generally that means that there should be a clear trough between an enhancement and the equatorial ionization anomaly (EIA).
-confined and have a border of lower TEC values (Δ≥20%) no farther than in 40° in longitude from the location of maximal TEC value. Mainly that means that we do not consider longitudinally stretched enhancements assuming different mechanism of their generation.
These criteria were applied to analysis of global ionospheric maps (GIMs). Currently, these maps are provided by several scientific groups: CODE (codg), ESA (esag), JPL (jplg), UPC (upcg), Whuan university (whug), Chinese Academy of Sciences (CAS -casg). IGS service also provides maps (igsg) created as a combination of maps from CODE, UPS, ESA and JPL. The spatial resolution is 2.5°x5° in latitude and longitude, respectively, and temporal one is 2  In the present paper SWARM in-situ measurements of electron density are used.
The project COSMIC (Constellation Observing System for Meteorology Ionosphere & Climate) provide measurements of upper atmosphere and ionosphere parameters. In the present paper we use TEC profiles obtained via radio occultation (RO) receiving of GPS signals. To distinguish this data from the standard ground-based TEC measurements, we use abbreviation SS TEC (satellite-to-satellite TEC). COSMIC data is freely provided as NetCDF files (https://cdaac-www.cosmic.ucar.edu/).
We analyze mainly the occurrence rate of LTE and its dependence on space weather. Quantitative analysis of LTEs generally consists of definition of maximal TEC value over the investigation region and calculation of its relation to mean TEC value over the region. Analysis of the dependence of these parameters on near space conditions was made during the investigation. LTE shapes vary widely and are quite difficult for formalization.
To analyse connection of the observed features of ionospheric dynamics with geomagnetic field we use SuperDARN altitude adjusted corrected geomagnetic coordinates (AACGM) (Shepherd, 2014)   It is possible to distinguish two parts in presented LTE: midlatitudinal (MLTE) and subpolar (SLTE). The LTE of April 5 has strong subpolar part and weaker but still pronounced midlatitudinal one. As it will be shown later, such a strong SLTE is not typical and in some cases it is not detected at all. However, both MLTE and SLTE were presented during this even quite clearly for several hours and that was the main reason to describe this particular case in more details. It is necessary to say that the LTEs are detected most clearly over Atlantic and Indian oceans, where amount of GNSS stations is insufficient. White squares in fig. 1 mark location of the receivers providing CODE with data for TEC maps. Only a few are located in ocean (on islands) and the only is in 30-60°S latitudes of Indian Ocean (Kerguelen Islands, KERG).

An example of a clearly observed LTE was detected at
Therefore LTE detection has to be confirmed by other observations.
In-situ measurements of electron concentration Ne from SWARM satellites allow us to validate TEC distribution presented by GIM. Left panel of figure 2 presents Ne values, observed during 8-14 UT at April 5, 2014. Each track is marked by a colored dot corresponding to satellite: Alpha (red), Bravo (blue) and Charlie (cyan); digits of the corresponding color marks the satellite position at the the beginning and the end of the track in a format HHMM (hours and minutes). All the satellites were moving from equator to pole.
The area of extremely high concentration of electrons is clearly observed in data from all the three satellites. Blank areas in measurements from Alpha and Charlie during 11:30-13:30 mark the zone of concentrations exceeding color axis limitation.
Temporal differences between passages of the satellites allow us to observe the dynamics of the LTE. The most intensive part is shown by Alpha's measurements. Charlie is ahead of Alpha by about 15 minutes and 2.5° of longitude and its measurements in general show lower concentration especially for a period 8-12 UT. Most probably such a difference is caused by movement of the enhancement: according to the GIM LTE is located in subsolar region and follows the Sun.
Bravo is about 30 min and 12° behind Alpha and its measurements shows significantly lower concentration than the other satellites. It could point not only to the disturbance displacement but also to its distribution with altitude, since the orbit of Bravo is 70 km higher than those of Alpha and Charlie.
The distribution of electron concentration with altitude can be analyzed using radio occultation measurements by COSMIC satellites. Profiles of SS TEC during April 5 are presented in the right panel of fig. 2. Each SS TEC value in a profile is obtained on a bent satellite-to-satellite ray and is attributed to a tangent point of the ray (Rocken et al., 2000). It is quite clear that the detected disturbance was propagating according to solar motion and had the highest electron concentration in F region at about 11 UT. Profiles also show that electron concentration at an altitude 460 km could be 1.5-2 times higher that at 530 km, which is in a correspondence with SWARM measurements.
LTEs similar to the one detected on April 5 could be observed during several days in a row. In the particular case of April 2014, LTEs southward of Africa were detected since March 18 till April 11. TEC maps at 10:00 UT for April 1-9, 2014 are presented at the left side of fig. 3. The geomagnetic conditions during this period were slightly disturbed: maximal value of Kp was 4 (April 7), and minimal Dst value was about -25 nT (April 7-8). The intensity and shape of the presented LTEs vary from day to day, but at the same UT all of the LTEs occupy the same region. Intensities of MLTE and SLTE vary independently. SLTE is more intense only on April 5. Mostly its intensity is either close to that of MLTE (April 1,3,4,7,9) or lower (April 2, 6 and 8). We define such a continuous sequence of LTEs observed day by day as  . 4, top). It is possible to see that only several short gaps separate this series from two others in autumn and probably the entire period of late March-July should be considered to include one long series. Such a long sequence occupying the third part of a year definitely points to some regular process.
For the other years the same season contains majority of the LTE series, but separated with more frequent and wider gaps. It is interesting to see that during a year of low solar activity (2018) we detect more series than during a moderately active one Being observed separately SH LTEs were previously supposed to be a relatively rare phenomenon produced by some specific condition of near space (Edemskiy et al., 2018). However the data presented above showed that LTEs occur quite often and can be observed in a sequence during a relatively long period when geomagnetic conditions and solar parameters vary significantly. The presented distributions did not reveal any pronounced dependence except the one between maximal TEC value in the region and solar flux intensity ( fig. 5a). Obviously TECmax linearly depends on total amount of electrons in ionosphere or global electron content and the last one is known to be dependent on F10.7 index (e.g., Astafyeva et al., 2008). At the same time it is surprising that the other distributions in fig. 5  The figure shows that most of the bright SLTEs were detected at the moments of negative Dst and SYM-H, and high values of AE index. In total that means that bright SLTEs are often observed during disturbed geomagnetic conditions. It is known that SEDs generated in high latitudes during geomagnetic storms could be observed in TEC maps as localized enhancements (e.g. Foster and and Rideout, 2007) and the detected SLTE could be a manifestation of some SED.
According to Foster (2008) SEDs are typically observed during severe geomagnetic storms and generally are formed by a Fregion plasma driven upward and poleward (ExB direction) by eastward electric field penetrated into the inner magnetosphere at the early phase of a geomagnetic storm. Being formed by the fountain effect the enhanced plasma of EIA peaks can be redistributed during extreme events when uplifting plasma reaches higher-latitude flux tubes, resulting in enhanced electron density near the plasmapause. Most often such uplifts are observed in the dusk sector (Foster, 2008).
Further development of the event can lead to generation of sub-auroral polarization stream creating SED as a connection between dusk sector and a region of dayside cusp. So partially the detected SLTEs could be generated via the described mechanism.
At the same time several features of SLTE should be highlighted. First, intense SLTEs were detected during a relatively quiet period as well. At least a quarter of them were detected with AE index values lower that 200 nT ( fig. 6a). Second, mostly SEDs are believed to be plume-shaped, clearly connected to EIA region, and have high intensities along the entire plume. The used criteria excluded from consideration both the stretched formations and the ones having connection to EIA.
So not all of the SLTEs are produced by some kind of SEDs and even if they are, the mechanism of their generation should differs from the one in NH. Midlatitudinal LTEs are mostly detected in the same region of SH (at 10 UT), but demonstrate wide variety of shapes. It is difficult to say that their generation is driven by space weather since no clear dependence on its main parameters were found for both the occurrence rate and the intensities of LTEs. Most probable the mechanism of their formation is connected to some kind of plasma redistribution since the most often the enhancements are observed during autumn-winter period (April-August, fig 4) when intensity of solar ionization in middle latitudes should be less effective than during summer. Apparently the mechanism is not connected with or not organized like the fountain effect since last one typically gives a quasisymmetrical (with respect to equator) pattern, and similar LTEs were not detected in magneto-conjugated region of NH.
Moreover the intensities of TEC in corresponding part on NH during LTE detection are typically lower than ones in SH.
Together with seasonal asymmetry that reminds winter anomaly (WA) phenomenon: F2-layer density values are greater in the winter hemisphere than in the summer hemisphere. It should be noted that, using COSMIC RO data Gowtam and Tulasi Ram (2017) showed that at altitudes within 300-700 km WA effect is confined only to morning-noon hours and only to lowlatitudes, claiming absence of WA in middle latitudes . Yasyukevich et. al. (2018) analysing GIM and satellites' data confirmed that SH WA is much less pronounced than NH WA and the region of its observation is mostly located in the southern part of Indian Ocean. The authors also showed dependence of the anomaly intensity on solar activity and claimed that it could be observed only during high solar activity years. Moreover, they concluded that in TEC the anomaly could be observed only in periods with F10.7 > 170 SFU. As it was shown above only intensity of LTE depends on F10.7 , but not the of particle flux located within L=2.5-3 in South Atlantic. The position of the plume was in a good correlation with a typical LTE position. However, the plume was observed in December when occurrence rate is minimal ( fig. 4). Moreover, for the LTE analyzed in detail by Edemskiy et al. (2018) it was shown that particle precipitations are not responsible for that LTE.
So most probably LTEs are not directly connected with the increased fluxes.
Statistically electron concentration over the western part of Indian Ocean is enhanced during equinox periods. Jee et al. As a conclusion we should say that at the present moment the generation mechanism is still unclear for us. The phenomenon of SH LTE is observed quite regularly, in periods of different solar activity and under different conditions of near space manifesting itself even during geomagnetically quiet periods. Since we did not detect symmetrical phenomena in Northern Hemisphere we could conclude that the enhancements are a feature of Southern Hemisphere ionosphere and therefore they should be driven by combination of its specific conditions: geomagnetic field, oceanic ionosphere and system of winds.
Such a regular phenomenon should be taken into account by models as well. Currently it is difficult to say if it is reproduced by models, since it is not well described in the literature. We could mention a paper by Lee et al. (2011) who showed presence of enhanced electron concentration formation over western part of Indian Ocean using measurements from GRACE and CHAMP satellites. The authors concluded that 2001 and 2007 IRI models did not predict the observed enhancement at all. So the phenomenon should be investigated more precisely since it will surely give us more clear understanding of global distribution of ionospheric plasma.

Conclusions
The paper shows that localized TEC enhancements in Southern Hemisphere are observed quite regularly and can be detected The presented data lead us to the opinion that despite the observed LTEs were supposed to be an ionospheric disturbance they most probable are a feature of the Southern Hemisphere ionosphere. The phenomenon should be investigated in more details with some additional methods including comparison with different models of ionosphere.